Low power decentralized differentially private multi-armed bandit algorithm based performance improvement on long-range radio network

Low power wide area networks are considerably focused to consume minimum power though increasing a communication array for end policies. Numerous methods were proposed for enhancing the functioning of Long-Range Radio (LoRa) network, but few methods failed to achieve significant performances. The configuration of LoRa networks is very challenging one, especially when increasing the number of nodes, the data rate is affected. This leads to higher collision rates and lower packet delivery ratios, particularly in scenarios with high node densities or interference. Additionally, methods that do not employ adaptive data rate strategies fail to optimize transmission parameters like coding rate and transmission power dynamically. This results in sub optimal throughput performance, especially during periods of high network activity. Furthermore, inefficient handling of radio-related parameters like carrier frequency and bandwidth contributes to decreased packet delivery ratio and overall network performance. To tackle this, a Low power Decentralized differentially private Multi-armed bandit Algorithm (LDMA) is proposed to enhance the performance of network while occupying large area transmission (upto 10 km). The designed algorithm works as a two phase to construct the transmission parameters based on the acknowledgement messages. The complete experiment is simulated through MATLAB. The results prove that proposed model attains minimum energy consumption of 7.82 J (56.3%) when compared to straightforward bounding technique, maximum packet delivery ratio of 76.3%, lower transmission delay of 0.082 ms and path loss of 178 dB and symbol period 0.042 Ts attained, leads to achieve higher transmission rate, which is better than the existing approaches.